{
  "id": 67018,
  "title": "Official External Data Thread",
  "url": "/competitions/quickdraw-doodle-recognition/discussion/67018",
  "author_name": "Julia Elliott",
  "post_date": "2018-09-27T18:44:45.939000",
  "votes": 7,
  "comment_count": 39,
  "views": 0,
  "content": "<p>Per the competition's rules, this competition allows the use of <strong>publicly, freely available external data</strong>. The source of any external data must be posted to this thread prior to the Entry Deadline.</p>",
  "messages": [
    {
      "id": 394972,
      "postDate": "2018-09-27T18:44:45.940Z",
      "content": "<p>Per the competition's rules, this competition allows the use of <strong>publicly, freely available external data</strong>. The source of any external data must be posted to this thread prior to the Entry Deadline.</p>",
      "rawMarkdown": "Per the competition's rules, this competition allows the use of **publicly, freely available external data**. The source of any external data must be posted to this thread prior to the Entry Deadline.",
      "votes": 7
    },
    {
      "id": 395044,
      "postDate": "2018-09-27T21:33:03.560Z",
      "content": "<p>I'm using Google's hosted dataset at <a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a> . I don't know if the site has anything beyond what we're already given so just in case. Plus it's a good resource for others to see on this thread.</p>",
      "rawMarkdown": "I'm using Google's hosted dataset at https://github.com/googlecreativelab/quickdraw-dataset . I don't know if the site has anything beyond what we're already given so just in case. Plus it's a good resource for others to see on this thread.",
      "votes": 8
    },
    {
      "id": 402392,
      "postDate": "2018-10-11T16:20:31.243Z",
      "content": "<p>models from:</p>\n\n<p><a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>\n\n<p><a href=\"https://github.com/ansleliu/LightNet\">https://github.com/ansleliu/LightNet</a></p>\n\n<p><a href=\"https://github.com/fyu/drn\">https://github.com/fyu/drn</a></p>\n\n<p><a href=\"https://github.com/mandeer/Classifier\">https://github.com/mandeer/Classifier</a> </p>\n\n<p><a href=\"https://github.com/mapillary/inplace_abn\">https://github.com/mapillary/inplace_abn</a></p>\n\n<p>external dataset:</p>\n\n<p><a href=\"http://cybertron.cg.tu-berlin.de/eitz/projects/classifysketch/\">http://cybertron.cg.tu-berlin.de/eitz/projects/classifysketch/</a></p>\n\n<p><a href=\"http://cybertron.cg.tu-berlin.de/eitz/projects/sbsr/\">http://cybertron.cg.tu-berlin.de/eitz/projects/sbsr/</a></p>\n\n<p><a href=\"http://sketchy.eye.gatech.edu/explore/rabbit.html\">http://sketchy.eye.gatech.edu/explore/rabbit.html</a></p>\n\n<p><a href=\"http://multimediauts.org/sketch/dataset.html\">http://multimediauts.org/sketch/dataset.html</a></p>\n\n<p><a href=\"https://www.eecs.qmul.ac.uk/~qian/Project_cvpr16.html\">https://www.eecs.qmul.ac.uk/~qian/Project_cvpr16.html</a></p>\n\n<p><a href=\"https://panly099.github.io/crssdomain.html\">https://panly099.github.io/crssdomain.html</a></p>\n\n<p><a href=\"https://www.cc.gatech.edu/~hays/7476/projects/Cusuh/\">https://www.cc.gatech.edu/~hays/7476/projects/Cusuh/</a></p>\n\n<p><a href=\"http://sketchx.eecs.qmul.ac.uk/downloads/\">http://sketchx.eecs.qmul.ac.uk/downloads/</a></p>\n\n<p><a href=\"https://github.com/panly099/sketchSynthesis\">https://github.com/panly099/sketchSynthesis</a></p>\n\n<p>Human Object Sketches:Datasets, Descriptors, Computational Recognition and 3d Shape Retrieval\n(<a href=\"https://d-nb.info/1029192820/34\">https://d-nb.info/1029192820/34</a>)</p>\n\n<p><a href=\"https://github.com/vlandham/quickdraw/blob/master/analysis/all_aggregates.csv\">https://github.com/vlandham/quickdraw/blob/master/analysis/all_aggregates.csv</a></p>\n\n<p><a href=\"http://val.serc.iisc.ernet.in/eotd/epipartviz/\">http://val.serc.iisc.ernet.in/eotd/epipartviz/</a></p>\n\n<p>Sketchyscene <a href=\"http://openaccess.thecvf.com/content_ECCV_2018/papers/Changqing_Zou_SketchyScene_Richly-Annotated_Scene_ECCV_2018_paper.pdf\">http://openaccess.thecvf.com/content_ECCV_2018/papers/Changqing_Zou_SketchyScene_Richly-Annotated_Scene_ECCV_2018_paper.pdf</a></p>",
      "rawMarkdown": "models from:\n\nhttps://github.com/Cadene/pretrained-models.pytorch\n\nhttps://github.com/ansleliu/LightNet\n\nhttps://github.com/fyu/drn\n\nhttps://github.com/mandeer/Classifier \n\nhttps://github.com/mapillary/inplace_abn\n\nexternal dataset:\n\nhttp://cybertron.cg.tu-berlin.de/eitz/projects/classifysketch/\n\nhttp://cybertron.cg.tu-berlin.de/eitz/projects/sbsr/\n\nhttp://sketchy.eye.gatech.edu/explore/rabbit.html\n\nhttp://multimediauts.org/sketch/dataset.html\n\nhttps://www.eecs.qmul.ac.uk/~qian/Project_cvpr16.html\n\nhttps://panly099.github.io/crssdomain.html\n\nhttps://www.cc.gatech.edu/~hays/7476/projects/Cusuh/\n\nhttp://sketchx.eecs.qmul.ac.uk/downloads/\n\nhttps://github.com/panly099/sketchSynthesis\n\nHuman Object Sketches:Datasets, Descriptors, Computational Recognition and 3d Shape Retrieval\n(https://d-nb.info/1029192820/34)\n\nhttps://github.com/vlandham/quickdraw/blob/master/analysis/all_aggregates.csv\n\nhttp://val.serc.iisc.ernet.in/eotd/epipartviz/\n\nSketchyscene http://openaccess.thecvf.com/content_ECCV_2018/papers/Changqing_Zou_SketchyScene_Richly-Annotated_Scene_ECCV_2018_paper.pdf",
      "votes": 5
    },
    {
      "id": 402895,
      "postDate": "2018-10-12T14:32:09.647Z",
      "content": "<p>keras pre-trained:</p>\n\n<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "keras pre-trained:\n\n[https://github.com/keras-team/keras-applications][1]\n\n\n  [1]: https://github.com/keras-team/keras-applications",
      "votes": 3,
      "replies": [
        {
          "id": 408098,
          "postDate": "2018-10-22T10:09:10.187Z",
          "content": "<p>I use keras pre-trained models too </p>",
          "rawMarkdown": "I use keras pre-trained models too "
        }
      ]
    },
    {
      "id": 420172,
      "postDate": "2018-11-13T07:37:50.030Z",
      "content": "<p>Keras pre-trained models from:\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\n<a href=\"https://github.com/fchollet/deep-learning-models\">https://github.com/fchollet/deep-learning-models</a>\n<a href=\"https://github.com/qubvel/classification_models\">https://github.com/qubvel/classification_models</a>\nNo external dataset.</p>",
      "rawMarkdown": "Keras pre-trained models from:\nhttps://github.com/keras-team/keras-applications\nhttps://github.com/fchollet/deep-learning-models\nhttps://github.com/qubvel/classification_models\nNo external dataset.",
      "votes": 1
    },
    {
      "id": 408083,
      "postDate": "2018-10-22T09:36:57.813Z",
      "content": "<p>Pre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>\n\n<p>External Dataset: \n<a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a></p>",
      "rawMarkdown": "Pre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\n\nExternal Dataset: \nhttps://github.com/googlecreativelab/quickdraw-dataset",
      "votes": 1
    },
    {
      "id": 432225,
      "postDate": "2018-12-03T15:23:10.567Z",
      "content": "<p><a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>",
      "rawMarkdown": "https://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch"
    },
    {
      "id": 432181,
      "postDate": "2018-12-03T14:19:35.930Z",
      "content": "<p>models from:\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>",
      "rawMarkdown": "models from:\nhttps://github.com/Cadene/pretrained-models.pytorch"
    },
    {
      "id": 432116,
      "postDate": "2018-12-03T12:34:40.473Z",
      "content": "<p>keras pre-trained:</p>\n\n<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "keras pre-trained:\n\nhttps://github.com/keras-team/keras-applications"
    },
    {
      "id": 431885,
      "postDate": "2018-12-03T03:57:04.733Z",
      "content": "<p>Keras pre-trained models\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "Keras pre-trained models\nhttps://github.com/keras-team/keras-applications"
    },
    {
      "id": 431552,
      "postDate": "2018-12-02T13:37:31.323Z",
      "content": "<p>Keras models from:\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "Keras models from:\nhttps://github.com/keras-team/keras-applications"
    },
    {
      "id": 430021,
      "postDate": "2018-11-29T17:03:01.803Z",
      "content": "<p>models from:</p>\n\n<p><a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>",
      "rawMarkdown": "models from:\n\nhttps://github.com/Cadene/pretrained-models.pytorch"
    },
    {
      "id": 429967,
      "postDate": "2018-11-29T15:56:48.273Z",
      "content": "<p>Pre-trained models:-\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a></p>",
      "rawMarkdown": "Pre-trained models:-\nhttps://pytorch.org/docs/stable/torchvision/models.html"
    },
    {
      "id": 429889,
      "postDate": "2018-11-29T14:10:44.853Z",
      "content": "<p>used model from (not pretrained)\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\n<a href=\"https://github.com/tensorflow/models\">https://github.com/tensorflow/models</a></p>",
      "rawMarkdown": "used model from (not pretrained)\nhttps://github.com/keras-team/keras-applications\nhttps://github.com/tensorflow/models"
    },
    {
      "id": 429872,
      "postDate": "2018-11-29T13:42:38.140Z",
      "content": "<p>Tensorflow models\n<a href=\"https://github.com/tensorflow/models\">https://github.com/tensorflow/models</a></p>",
      "rawMarkdown": "Tensorflow models\nhttps://github.com/tensorflow/models"
    },
    {
      "id": 428615,
      "postDate": "2018-11-27T15:30:31.343Z",
      "content": "<p>Keras pre-trained models from:\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\nPre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>\n<a href=\"https://github.com/tonylins/pytorch-mobilenet-v2\">https://github.com/tonylins/pytorch-mobilenet-v2</a>\n<a href=\"https://github.com/KaiyangZhou/deep-person-reid\">https://github.com/KaiyangZhou/deep-person-reid</a></p>",
      "rawMarkdown": "Keras pre-trained models from:\nhttps://github.com/keras-team/keras-applications\nPre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/tonylins/pytorch-mobilenet-v2\nhttps://github.com/KaiyangZhou/deep-person-reid"
    },
    {
      "id": 428173,
      "postDate": "2018-11-26T21:46:59.493Z",
      "content": "<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\n<a href=\"https://github.com/fchollet/deep-learning-models\">https://github.com/fchollet/deep-learning-models</a>\n<a href=\"https://github.com/qubvel/classification_models\">https://github.com/qubvel/classification_models</a></p>",
      "rawMarkdown": "https://github.com/keras-team/keras-applications\nhttps://github.com/fchollet/deep-learning-models\nhttps://github.com/qubvel/classification_models"
    },
    {
      "id": 427617,
      "postDate": "2018-11-25T21:26:58.930Z",
      "content": "<p>I made a copy of \"The UJIpenchars Database: Pen-Based Database with More Than 11K Isolated Handwritten Characters\" <a href=\"https://www.kaggle.com/sorokin/ujipenchars2\">https://www.kaggle.com/sorokin/ujipenchars2</a> and a simple kernel for parsing this dataset <a href=\"https://www.kaggle.com/sorokin/convert-ujipenchars2-to-the-quick-draw-dataset\">https://www.kaggle.com/sorokin/convert-ujipenchars2-to-the-quick-draw-dataset</a></p>",
      "rawMarkdown": "I made a copy of \"The UJIpenchars Database: Pen-Based Database with More Than 11K Isolated Handwritten Characters\" https://www.kaggle.com/sorokin/ujipenchars2 and a simple kernel for parsing this dataset https://www.kaggle.com/sorokin/convert-ujipenchars2-to-the-quick-draw-dataset"
    },
    {
      "id": 427517,
      "postDate": "2018-11-25T17:19:15.683Z",
      "content": "<p>fastai pretrained models: <a href=\"https://github.com/fastai/fastai\">https://github.com/fastai/fastai</a></p>",
      "rawMarkdown": "fastai pretrained models: https://github.com/fastai/fastai"
    },
    {
      "id": 427499,
      "postDate": "2018-11-25T16:05:28.850Z",
      "content": "<p>I'm using keras pre-trained model:</p>\n\n<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "I'm using keras pre-trained model:\n\nhttps://github.com/keras-team/keras-applications"
    },
    {
      "id": 426032,
      "postDate": "2018-11-22T13:24:51.577Z",
      "content": "<p>I use pretrain model from here: </p>\n\n<p><a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>\n\n<p>Didn't use any external data.</p>",
      "rawMarkdown": "I use pretrain model from here: \n\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/keras-team/keras-applications\n\nDidn't use any external data."
    },
    {
      "id": 425095,
      "postDate": "2018-11-21T06:06:28.033Z",
      "content": "<p>Pre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>\nNo external Data</p>",
      "rawMarkdown": "Pre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\nNo external Data"
    },
    {
      "id": 424659,
      "postDate": "2018-11-20T14:02:06.883Z",
      "content": "<p><a href=\"https://github.com/qubvel/classification_models\">https://github.com/qubvel/classification_models</a></p>",
      "rawMarkdown": "https://github.com/qubvel/classification_models"
    },
    {
      "id": 424485,
      "postDate": "2018-11-20T07:23:50.597Z",
      "content": "<p>word2vec pretrained model:\n<a href=\"https://github.com/mmihaltz/word2vec-GoogleNews-vectors\">https://github.com/mmihaltz/word2vec-GoogleNews-vectors</a></p>",
      "rawMarkdown": "word2vec pretrained model:\nhttps://github.com/mmihaltz/word2vec-GoogleNews-vectors"
    },
    {
      "id": 423875,
      "postDate": "2018-11-19T07:16:08.477Z",
      "content": "<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "https://github.com/keras-team/keras-applications"
    },
    {
      "id": 422427,
      "postDate": "2018-11-16T07:49:58.237Z",
      "content": "<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "https://github.com/keras-team/keras-applications"
    },
    {
      "id": 421322,
      "postDate": "2018-11-14T22:45:14.327Z",
      "content": "<p><a href=\"https://github.com/tensorflow/magenta/tree/master/magenta/models/sketch_rnn\">https://github.com/tensorflow/magenta/tree/master/magenta/models/sketch_rnn</a></p>",
      "rawMarkdown": "https://github.com/tensorflow/magenta/tree/master/magenta/models/sketch_rnn"
    },
    {
      "id": 421316,
      "postDate": "2018-11-14T22:36:37.073Z",
      "content": "<p>quickdraw dataset\n<a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a></p>\n\n<p>pretrained pytorch resnets using imagenet</p>",
      "rawMarkdown": "quickdraw dataset\nhttps://github.com/googlecreativelab/quickdraw-dataset\n\npretrained pytorch resnets using imagenet"
    },
    {
      "id": 419207,
      "postDate": "2018-11-11T14:14:52.683Z",
      "content": "<p><a href=\"https://developers.google.com/public-data/docs/canonical/countries_csv\">https://developers.google.com/public-data/docs/canonical/countries_csv</a>\n<a href=\"https://opendata.socrata.com/dataset/Country-List-ISO-3166-Codes-Latitude-Longitude/mnkm-8ram/data\">https://opendata.socrata.com/dataset/Country-List-ISO-3166-Codes-Latitude-Longitude/mnkm-8ram/data</a></p>",
      "rawMarkdown": "https://developers.google.com/public-data/docs/canonical/countries_csv\nhttps://opendata.socrata.com/dataset/Country-List-ISO-3166-Codes-Latitude-Longitude/mnkm-8ram/data"
    },
    {
      "id": 417888,
      "postDate": "2018-11-09T00:04:20.960Z",
      "content": "<p>Pre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>\n\n<p>External Dataset: \n<a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a></p>",
      "rawMarkdown": "Pre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\n\nExternal Dataset: \nhttps://github.com/googlecreativelab/quickdraw-dataset"
    },
    {
      "id": 415770,
      "postDate": "2018-11-05T17:02:34.167Z",
      "content": "<p>keras pre-trained:</p>\n\n<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "rawMarkdown": "keras pre-trained:\n\nhttps://github.com/keras-team/keras-applications"
    },
    {
      "id": 414473,
      "postDate": "2018-11-02T20:35:54.050Z",
      "content": "<p><a href=\"https://keras.io/applications/#mobilenetv2\">https://keras.io/applications/#mobilenetv2</a></p>",
      "rawMarkdown": "https://keras.io/applications/#mobilenetv2"
    },
    {
      "id": 413956,
      "postDate": "2018-11-01T21:30:22.987Z",
      "content": "<p>pretrained models from <a href=\"https://github.com/tonylins/pytorch-mobilenet-v2\">https://github.com/tonylins/pytorch-mobilenet-v2</a></p>",
      "rawMarkdown": "pretrained models from https://github.com/tonylins/pytorch-mobilenet-v2"
    },
    {
      "id": 413335,
      "postDate": "2018-10-31T18:16:31.010Z",
      "content": "<p>pretrained vgg 16 <a href=\"https://github.com/machrisaa/tensorflow-vgg\">https://github.com/machrisaa/tensorflow-vgg</a></p>",
      "rawMarkdown": "pretrained vgg 16 https://github.com/machrisaa/tensorflow-vgg\n"
    },
    {
      "id": 411191,
      "postDate": "2018-10-27T15:33:21.583Z",
      "content": "<p>pretrained models available in the fastai library</p>",
      "rawMarkdown": "pretrained models available in the fastai library\n"
    },
    {
      "id": 411088,
      "postDate": "2018-10-27T11:32:50.683Z",
      "content": "<p>CNTK Pre-trained Image Models <a href=\"https://github.com/Microsoft/CNTK/blob/master/PretrainedModels/Image.md\">https://github.com/Microsoft/CNTK/blob/master/PretrainedModels/Image.md</a></p>",
      "rawMarkdown": "CNTK Pre-trained Image Models https://github.com/Microsoft/CNTK/blob/master/PretrainedModels/Image.md"
    },
    {
      "id": 409594,
      "postDate": "2018-10-24T14:46:47.457Z",
      "content": "<p>pretrained models available in the <a href=\"https://github.com/fastai/fastai\">fastai library</a></p>",
      "rawMarkdown": "pretrained models available in the [fastai library][1]\n\n\n  [1]: https://github.com/fastai/fastai"
    },
    {
      "id": 429970,
      "postDate": "2018-11-29T15:59:28.320Z",
      "rawMarkdown": "",
      "isDeleted": true
    },
    {
      "id": 428083,
      "postDate": "2018-11-26T18:18:22.513Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 395044,
      "author_name": "JohnM",
      "author_url": "",
      "post_date": "2018-09-27T21:33:03.560000",
      "content": "<p>I'm using Google's hosted dataset at <a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a> . I don't know if the site has anything beyond what we're already given so just in case. Plus it's a good resource for others to see on this thread.</p>",
      "votes": 8,
      "replies": []
    },
    {
      "id": 402392,
      "author_name": "hengck23",
      "author_url": "",
      "post_date": "2018-10-11T16:20:31.243000",
      "content": "<p>models from:</p>\n\n<p><a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>\n\n<p><a href=\"https://github.com/ansleliu/LightNet\">https://github.com/ansleliu/LightNet</a></p>\n\n<p><a href=\"https://github.com/fyu/drn\">https://github.com/fyu/drn</a></p>\n\n<p><a href=\"https://github.com/mandeer/Classifier\">https://github.com/mandeer/Classifier</a> </p>\n\n<p><a href=\"https://github.com/mapillary/inplace_abn\">https://github.com/mapillary/inplace_abn</a></p>\n\n<p>external dataset:</p>\n\n<p><a href=\"http://cybertron.cg.tu-berlin.de/eitz/projects/classifysketch/\">http://cybertron.cg.tu-berlin.de/eitz/projects/classifysketch/</a></p>\n\n<p><a href=\"http://cybertron.cg.tu-berlin.de/eitz/projects/sbsr/\">http://cybertron.cg.tu-berlin.de/eitz/projects/sbsr/</a></p>\n\n<p><a href=\"http://sketchy.eye.gatech.edu/explore/rabbit.html\">http://sketchy.eye.gatech.edu/explore/rabbit.html</a></p>\n\n<p><a href=\"http://multimediauts.org/sketch/dataset.html\">http://multimediauts.org/sketch/dataset.html</a></p>\n\n<p><a href=\"https://www.eecs.qmul.ac.uk/~qian/Project_cvpr16.html\">https://www.eecs.qmul.ac.uk/~qian/Project_cvpr16.html</a></p>\n\n<p><a href=\"https://panly099.github.io/crssdomain.html\">https://panly099.github.io/crssdomain.html</a></p>\n\n<p><a href=\"https://www.cc.gatech.edu/~hays/7476/projects/Cusuh/\">https://www.cc.gatech.edu/~hays/7476/projects/Cusuh/</a></p>\n\n<p><a href=\"http://sketchx.eecs.qmul.ac.uk/downloads/\">http://sketchx.eecs.qmul.ac.uk/downloads/</a></p>\n\n<p><a href=\"https://github.com/panly099/sketchSynthesis\">https://github.com/panly099/sketchSynthesis</a></p>\n\n<p>Human Object Sketches:Datasets, Descriptors, Computational Recognition and 3d Shape Retrieval\n(<a href=\"https://d-nb.info/1029192820/34\">https://d-nb.info/1029192820/34</a>)</p>\n\n<p><a href=\"https://github.com/vlandham/quickdraw/blob/master/analysis/all_aggregates.csv\">https://github.com/vlandham/quickdraw/blob/master/analysis/all_aggregates.csv</a></p>\n\n<p><a href=\"http://val.serc.iisc.ernet.in/eotd/epipartviz/\">http://val.serc.iisc.ernet.in/eotd/epipartviz/</a></p>\n\n<p>Sketchyscene <a href=\"http://openaccess.thecvf.com/content_ECCV_2018/papers/Changqing_Zou_SketchyScene_Richly-Annotated_Scene_ECCV_2018_paper.pdf\">http://openaccess.thecvf.com/content_ECCV_2018/papers/Changqing_Zou_SketchyScene_Richly-Annotated_Scene_ECCV_2018_paper.pdf</a></p>",
      "votes": 5,
      "replies": []
    },
    {
      "id": 402895,
      "author_name": "outrunner",
      "author_url": "",
      "post_date": "2018-10-12T14:32:09.647000",
      "content": "<p>keras pre-trained:</p>\n\n<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "votes": 3,
      "replies": [
        {
          "id": 408098,
          "author_name": "Serigne ",
          "author_url": "",
          "post_date": "2018-10-22T10:09:10.187000",
          "content": "<p>I use keras pre-trained models too </p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 420172,
      "author_name": "seigato",
      "author_url": "",
      "post_date": "2018-11-13T07:37:50.030000",
      "content": "<p>Keras pre-trained models from:\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\n<a href=\"https://github.com/fchollet/deep-learning-models\">https://github.com/fchollet/deep-learning-models</a>\n<a href=\"https://github.com/qubvel/classification_models\">https://github.com/qubvel/classification_models</a>\nNo external dataset.</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 408083,
      "author_name": "LarryHuang",
      "author_url": "",
      "post_date": "2018-10-22T09:36:57.813000",
      "content": "<p>Pre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>\n\n<p>External Dataset: \n<a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a></p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 432225,
      "author_name": "bestfitting",
      "author_url": "",
      "post_date": "2018-12-03T15:23:10.567000",
      "content": "<p><a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 432181,
      "author_name": "Tim Wu",
      "author_url": "",
      "post_date": "2018-12-03T14:19:35.930000",
      "content": "<p>models from:\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 432116,
      "author_name": "HAN CHEN",
      "author_url": "",
      "post_date": "2018-12-03T12:34:40.473000",
      "content": "<p>keras pre-trained:</p>\n\n<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 431885,
      "author_name": "lippyt",
      "author_url": "",
      "post_date": "2018-12-03T03:57:04.733000",
      "content": "<p>Keras pre-trained models\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 431552,
      "author_name": "Niclas Stoltenberg",
      "author_url": "",
      "post_date": "2018-12-02T13:37:31.323000",
      "content": "<p>Keras models from:\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 430021,
      "author_name": "chicm",
      "author_url": "",
      "post_date": "2018-11-29T17:03:01.803000",
      "content": "<p>models from:</p>\n\n<p><a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 429967,
      "author_name": "Shekhar Jha",
      "author_url": "",
      "post_date": "2018-11-29T15:56:48.273000",
      "content": "<p>Pre-trained models:-\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 429889,
      "author_name": "hahaha",
      "author_url": "",
      "post_date": "2018-11-29T14:10:44.853000",
      "content": "<p>used model from (not pretrained)\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\n<a href=\"https://github.com/tensorflow/models\">https://github.com/tensorflow/models</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 429872,
      "author_name": "beluga",
      "author_url": "",
      "post_date": "2018-11-29T13:42:38.140000",
      "content": "<p>Tensorflow models\n<a href=\"https://github.com/tensorflow/models\">https://github.com/tensorflow/models</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 428615,
      "author_name": "JiaqiFan",
      "author_url": "",
      "post_date": "2018-11-27T15:30:31.343000",
      "content": "<p>Keras pre-trained models from:\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\nPre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>\n<a href=\"https://github.com/tonylins/pytorch-mobilenet-v2\">https://github.com/tonylins/pytorch-mobilenet-v2</a>\n<a href=\"https://github.com/KaiyangZhou/deep-person-reid\">https://github.com/KaiyangZhou/deep-person-reid</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 428173,
      "author_name": "Lukasz Grad",
      "author_url": "",
      "post_date": "2018-11-26T21:46:59.493000",
      "content": "<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a>\n<a href=\"https://github.com/fchollet/deep-learning-models\">https://github.com/fchollet/deep-learning-models</a>\n<a href=\"https://github.com/qubvel/classification_models\">https://github.com/qubvel/classification_models</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 427617,
      "author_name": "ivan",
      "author_url": "",
      "post_date": "2018-11-25T21:26:58.930000",
      "content": "<p>I made a copy of \"The UJIpenchars Database: Pen-Based Database with More Than 11K Isolated Handwritten Characters\" <a href=\"https://www.kaggle.com/sorokin/ujipenchars2\">https://www.kaggle.com/sorokin/ujipenchars2</a> and a simple kernel for parsing this dataset <a href=\"https://www.kaggle.com/sorokin/convert-ujipenchars2-to-the-quick-draw-dataset\">https://www.kaggle.com/sorokin/convert-ujipenchars2-to-the-quick-draw-dataset</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 427517,
      "author_name": "Alexander Schumacher",
      "author_url": "",
      "post_date": "2018-11-25T17:19:15.683000",
      "content": "<p>fastai pretrained models: <a href=\"https://github.com/fastai/fastai\">https://github.com/fastai/fastai</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 427499,
      "author_name": "Hoàng Tùng Lâm",
      "author_url": "",
      "post_date": "2018-11-25T16:05:28.850000",
      "content": "<p>I'm using keras pre-trained model:</p>\n\n<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 426032,
      "author_name": "wh1te",
      "author_url": "",
      "post_date": "2018-11-22T13:24:51.577000",
      "content": "<p>I use pretrain model from here: </p>\n\n<p><a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>\n<a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>\n\n<p>Didn't use any external data.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 425095,
      "author_name": "Mir Mustafa Ali",
      "author_url": "",
      "post_date": "2018-11-21T06:06:28.033000",
      "content": "<p>Pre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>\nNo external Data</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 424659,
      "author_name": "Guanshuo Xu",
      "author_url": "",
      "post_date": "2018-11-20T14:02:06.883000",
      "content": "<p><a href=\"https://github.com/qubvel/classification_models\">https://github.com/qubvel/classification_models</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 424485,
      "author_name": "Satoshi Tanaka",
      "author_url": "",
      "post_date": "2018-11-20T07:23:50.597000",
      "content": "<p>word2vec pretrained model:\n<a href=\"https://github.com/mmihaltz/word2vec-GoogleNews-vectors\">https://github.com/mmihaltz/word2vec-GoogleNews-vectors</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 423875,
      "author_name": "Ashish Lal",
      "author_url": "",
      "post_date": "2018-11-19T07:16:08.477000",
      "content": "<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 422427,
      "author_name": "TPloveYXT520",
      "author_url": "",
      "post_date": "2018-11-16T07:49:58.237000",
      "content": "<p><a href=\"https://github.com/keras-team/keras-applications\">https://github.com/keras-team/keras-applications</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 421322,
      "author_name": "Dmytro Danevskyi",
      "author_url": "",
      "post_date": "2018-11-14T22:45:14.327000",
      "content": "<p><a href=\"https://github.com/tensorflow/magenta/tree/master/magenta/models/sketch_rnn\">https://github.com/tensorflow/magenta/tree/master/magenta/models/sketch_rnn</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 421316,
      "author_name": "marii",
      "author_url": "",
      "post_date": "2018-11-14T22:36:37.073000",
      "content": "<p>quickdraw dataset\n<a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a></p>\n\n<p>pretrained pytorch resnets using imagenet</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 419207,
      "author_name": "Artyom Palvelev",
      "author_url": "",
      "post_date": "2018-11-11T14:14:52.683000",
      "content": "<p><a href=\"https://developers.google.com/public-data/docs/canonical/countries_csv\">https://developers.google.com/public-data/docs/canonical/countries_csv</a>\n<a href=\"https://opendata.socrata.com/dataset/Country-List-ISO-3166-Codes-Latitude-Longitude/mnkm-8ram/data\">https://opendata.socrata.com/dataset/Country-List-ISO-3166-Codes-Latitude-Longitude/mnkm-8ram/data</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 417888,
      "author_name": "Finlay",
      "author_url": "",
      "post_date": "2018-11-09T00:04:20.960000",
      "content": "<p>Pre-trained model:\n<a href=\"https://pytorch.org/docs/stable/torchvision/models.html\">https://pytorch.org/docs/stable/torchvision/models.html</a>\n<a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a></p>\n\n<p>External Dataset: \n<a href=\"https://github.com/googlecreativelab/quickdraw-dataset\">https://github.com/googlecreativelab/quickdraw-dataset</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 415770,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-11-05T17:02:34.167000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 414473,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-11-02T20:35:54.050000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 413956,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-11-01T21:30:22.987000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 413335,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-10-31T18:16:31.010000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 411191,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-10-27T15:33:21.583000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 411088,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-10-27T11:32:50.683000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 409594,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-10-24T14:46:47.457000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 429970,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-11-29T15:59:28.320000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 428083,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-11-26T18:18:22.513000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "394972": "Per the competition's rules, this competition allows the use of **publicly, freely available external data**. The source of any external data must be posted to this thread prior to the Entry Deadline.",
    "395044": "I'm using Google's hosted dataset at https://github.com/googlecreativelab/quickdraw-dataset . I don't know if the site has anything beyond what we're already given so just in case. Plus it's a good resource for others to see on this thread.",
    "402392": "models from:\n\nhttps://github.com/Cadene/pretrained-models.pytorch\n\nhttps://github.com/ansleliu/LightNet\n\nhttps://github.com/fyu/drn\n\nhttps://github.com/mandeer/Classifier \n\nhttps://github.com/mapillary/inplace_abn\n\nexternal dataset:\n\nhttp://cybertron.cg.tu-berlin.de/eitz/projects/classifysketch/\n\nhttp://cybertron.cg.tu-berlin.de/eitz/projects/sbsr/\n\nhttp://sketchy.eye.gatech.edu/explore/rabbit.html\n\nhttp://multimediauts.org/sketch/dataset.html\n\nhttps://www.eecs.qmul.ac.uk/~qian/Project_cvpr16.html\n\nhttps://panly099.github.io/crssdomain.html\n\nhttps://www.cc.gatech.edu/~hays/7476/projects/Cusuh/\n\nhttp://sketchx.eecs.qmul.ac.uk/downloads/\n\nhttps://github.com/panly099/sketchSynthesis\n\nHuman Object Sketches:Datasets, Descriptors, Computational Recognition and 3d Shape Retrieval\n(https://d-nb.info/1029192820/34)\n\nhttps://github.com/vlandham/quickdraw/blob/master/analysis/all_aggregates.csv\n\nhttp://val.serc.iisc.ernet.in/eotd/epipartviz/\n\nSketchyscene http://openaccess.thecvf.com/content_ECCV_2018/papers/Changqing_Zou_SketchyScene_Richly-Annotated_Scene_ECCV_2018_paper.pdf",
    "402895": "keras pre-trained:\n\n[https://github.com/keras-team/keras-applications][1]\n\n\n  [1]: https://github.com/keras-team/keras-applications",
    "420172": "Keras pre-trained models from:\nhttps://github.com/keras-team/keras-applications\nhttps://github.com/fchollet/deep-learning-models\nhttps://github.com/qubvel/classification_models\nNo external dataset.",
    "408083": "Pre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\n\nExternal Dataset: \nhttps://github.com/googlecreativelab/quickdraw-dataset",
    "432225": "https://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch",
    "432181": "models from:\nhttps://github.com/Cadene/pretrained-models.pytorch",
    "432116": "keras pre-trained:\n\nhttps://github.com/keras-team/keras-applications",
    "431885": "Keras pre-trained models\nhttps://github.com/keras-team/keras-applications",
    "431552": "Keras models from:\nhttps://github.com/keras-team/keras-applications",
    "430021": "models from:\n\nhttps://github.com/Cadene/pretrained-models.pytorch",
    "429967": "Pre-trained models:-\nhttps://pytorch.org/docs/stable/torchvision/models.html",
    "429889": "used model from (not pretrained)\nhttps://github.com/keras-team/keras-applications\nhttps://github.com/tensorflow/models",
    "429872": "Tensorflow models\nhttps://github.com/tensorflow/models",
    "428615": "Keras pre-trained models from:\nhttps://github.com/keras-team/keras-applications\nPre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/tonylins/pytorch-mobilenet-v2\nhttps://github.com/KaiyangZhou/deep-person-reid",
    "428173": "https://github.com/keras-team/keras-applications\nhttps://github.com/fchollet/deep-learning-models\nhttps://github.com/qubvel/classification_models",
    "427617": "I made a copy of \"The UJIpenchars Database: Pen-Based Database with More Than 11K Isolated Handwritten Characters\" https://www.kaggle.com/sorokin/ujipenchars2 and a simple kernel for parsing this dataset https://www.kaggle.com/sorokin/convert-ujipenchars2-to-the-quick-draw-dataset",
    "427517": "fastai pretrained models: https://github.com/fastai/fastai",
    "427499": "I'm using keras pre-trained model:\n\nhttps://github.com/keras-team/keras-applications",
    "426032": "I use pretrain model from here: \n\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/keras-team/keras-applications\n\nDidn't use any external data.",
    "425095": "Pre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\nNo external Data",
    "424659": "https://github.com/qubvel/classification_models",
    "424485": "word2vec pretrained model:\nhttps://github.com/mmihaltz/word2vec-GoogleNews-vectors",
    "423875": "https://github.com/keras-team/keras-applications",
    "422427": "https://github.com/keras-team/keras-applications",
    "421322": "https://github.com/tensorflow/magenta/tree/master/magenta/models/sketch_rnn",
    "421316": "quickdraw dataset\nhttps://github.com/googlecreativelab/quickdraw-dataset\n\npretrained pytorch resnets using imagenet",
    "419207": "https://developers.google.com/public-data/docs/canonical/countries_csv\nhttps://opendata.socrata.com/dataset/Country-List-ISO-3166-Codes-Latitude-Longitude/mnkm-8ram/data",
    "417888": "Pre-trained model:\nhttps://pytorch.org/docs/stable/torchvision/models.html\nhttps://github.com/Cadene/pretrained-models.pytorch\n\nExternal Dataset: \nhttps://github.com/googlecreativelab/quickdraw-dataset",
    "415770": "keras pre-trained:\n\nhttps://github.com/keras-team/keras-applications",
    "414473": "https://keras.io/applications/#mobilenetv2",
    "413956": "pretrained models from https://github.com/tonylins/pytorch-mobilenet-v2",
    "413335": "pretrained vgg 16 https://github.com/machrisaa/tensorflow-vgg\n",
    "411191": "pretrained models available in the fastai library\n",
    "411088": "CNTK Pre-trained Image Models https://github.com/Microsoft/CNTK/blob/master/PretrainedModels/Image.md",
    "409594": "pretrained models available in the [fastai library][1]\n\n\n  [1]: https://github.com/fastai/fastai",
    "429970": "",
    "428083": ""
  }
}